Impact of technology uncertainty on future low-carbon pathways in
, Ilkka Keppo, Marianne Zeyringer, Will Usher, Hannah Daly
UCL Energy Institute, University College London, Central House, 14 Upper Woburn Place, London, WC1H 0NN, United Kingdom
Received 19 June 2016
Received in revised form
18 September 2016
Accepted 25 September 2016
Available online 8 October 2016
Energy systems analysis
Energy and climate policy-making requires strong quantitative scientific evidence to devise robust and
consistent long-term decarbonisation strategies. Energy system modelling can provide crucial insights
into the inherent uncertainty in such strategies, which needs to be understood when designing appropriate policy measures.
This study contributes to the growing research area of uncertainty analysis in energy system models.
We combine consistent and realistic narratives on several technology dimensions with a global sensitivity analysis in a national, bottom-up, optimizing energy system model. This produces structured insights into the impact of low-carbon technology and resource availability on the long-term development
of the UK energy system under ambitious decarbonisation pathways. We explore a variety of result
metrics to present policy-relevant results in a useful and concise manner. The results provide valuable
information on the variability of fuel and technology use across the uncertainty space (e.g. a strong
variation in natural gas demand). We demonstrate the complementarities and substitutability of technologies (e.g. the dependency of hydrogen technologies on the availability of CCS). We highlight critical
low-carbon options and hedging strategies (e.g. the early decarbonisation of the electricity sector or the
stronger use of renewable sources as a hedging against failure in other technologies) and demonstrate
timing and path dependencies (e.g. the importance of early decarbonisation action in the presence of
multiple technology uncertainty). The results also show how the availability of a given technology can
have wider impacts elsewhere in the energy system, thus complicating the management of a long-term
Â© 2016 Elsevier Ltd. All rights reserved.
Quantitative energy modelling currently plays a fundamental
role in informing decision-making in energy and climate policies on
efficient long-term decarbonisation strategies, both on a global 
and national level . Given the uncertainty and complexity of
future low-carbon pathways, these energy-economic studies usually present their results as a small set of qualitatively different
scenarios which can be described as â€œplausible, challenging and
relevant stories about how the future might unfoldâ€ .
In brief, the modelling/policy process works as follows. Decision
makers rely upon policy reports for objective and balanced
information. The development of a policy report is supported by the
results of a modelling exercise. And these reports are used to help
set long-term target levels for emission reduction, energy efficiency
or use of renewable energies and outline the major technology
strategies to fulfil these objectives. But particularly when analysing
national policy reports, it becomes obvious that they usually rely on
a small set of scenarios (e.g. Refs. [19,34] derived from deterministic
energy system models. While acknowledging the need to deliver
clear and concise messages to policy makers, it is apparent that
such analyses are limited in terms of their description of uncertainty in the projected decarbonisation pathways they report. This
may lead to an overreliance on certain technologies or mitigation
strategies which feature strongly in the presented scenarios
While climate analysis has already progressed considerably in
terms of uncertainty analysis (cf. for example , it still seems to
be an emerging technique in energy systems studies. Different
approaches to represent uncertainty in energy-economic models
* Corresponding author.
E-mail addresses: [email protected] (B. Fais), [email protected] (I. Keppo), m.
[email protected] (M. Zeyringer), [email protected] (W. Usher), [email protected]
ucl.ac.uk (H. Daly).
Contents lists available at ScienceDirect
Energy Strategy Reviews
journal homepage: www.ees.elsevier.com/esr
2211-467X/Â© 2016 Elsevier Ltd. All rights reserved.
Energy Strategy Reviews 13-14 (2016) 154e168
can be observed in literature. The most common methods are
Sensitivity Analyses evaluating the variability of the model output as
a function of changing input parameters in deterministic models
. In order to further include interactions between input parameters, Global Sensitivity Techniques, which vary several uncertain input parameters at a time to explore the interaction effects, in
some cases through probabilistic and Monte Carlo methods, have
been developed . Recent studies with global sensitivity approaches in energy systems research are [1,2,6,73,89,92]. Other
methods include Stochastic Modelling [53,54,56,60,80,84,90],
Modelling to Generate Alternatives (MGA) [18,86,93] and Multi-model
Some background on these modelling techniques is provided in
Table A-1 in the Annex. Most of these advanced uncertainty
methods lead to a rising number of scenarios. This surely leads to a
better exploration of the uncertainty space, but at the same time it
has to be made sure that such studies produce relevant and
transparent policy insights .)
This study contributes to the growing research area of uncertainty analysis in energy system models. Using the approach of a
global sensitivity analysis in a national, bottom-up, optimizing
energy system model, the aim is to identify which low-carbon
technologies and resources have the most influence on the longterm development of the UK energy system under ambitious
decarbonisation pathways. Our motivation stems from the fact that
most forward looking scenarios rely on the rapid scaling up of
technologies, that currently either occupy a fairly small niche, but
have not yet demonstrated the capability of such growth or entered
commercial markets. While it seems likely that at least one of the
technologies will be able to scale up, it seems equally likely that at
least one of the technologies will suffer an unforeseen setback. Our
analysis aims to see how sensitive the outcomes are to the failure of
one or more key technologies, what are the interactions between
the technologies and at what point reaching targets may become
difficult. We emphasise the relevance to policy by (1) basing the
quantitative scenario analysis on consistent and, in the UK context,
realistic narratives for each technology dimension; (2) limiting the
analysis to a manageable number of scenarios such as to have
sufficient variability to assess the effect of technology uncertainty,
while still being able to analyse each scenario in detail and (3)
exploring various metrics to present the results across the scenario
matrix in an insightful and concise manner. The limited number of
dimensions of uncertainty allows us to conduct a global sensitivity
analysis by computing scenarios for the all the combinations of the
combinations of parameters.
The paper is structured as follows. Chapter 2 provides an overview of the methodology, including a description of the modelling
framework, the qualitative technology narratives and the approach
for the sensitivity analysis. The result metrics for the quantitative
scenario analysis are presented in Chapter 3 focusing on the
reference case, variability in fuel use, emissions and cost indicators
as well as insights on technology complementarity and substitutability. The paper concludes with a discussion of findings and policy
implications in Chapter 4.
2. Methodological approach
2.1. The national energy system model UKTM
We use the new national UK TIMES energy system model
(UKTM) [17,36] to conduct a quantitative scenario analysis. UKTM
has been developed at the UCL Energy Institute over the past two
years as the successor to the UK MARKAL model . It is based on
the model generator TIMES (The Integrated MARKAL-EFOM System), which is developed and maintained by the Energy
Technology Systems Analysis Programme (ETSAP) of the International Energy Agency (IEA) .
UKTM is a technology-oriented, dynamic, linear programming
optimisation model representing the entire UK energy system from
imports and domestic production of fuel resources, through fuel
processing and supply, explicit representation of infrastructures,
conversion to secondary energy carriers (including electricity, heat
and hydrogen), end-use technologies and energy service demands.
Generally, it minimizes the total welfare costs (under perfect
foresight) to meet exogenously defined sector energy demands
under a range of input assumptions and additional constraints. The
model delivers a cost optimal, system-wide solution for the energy
transition over the coming decades.
The model is divided into three supply side sectors (resources &
trade, processing & infrastructure and electricity generation) and
five demand sectors (residential, services, industry, transport and
agriculture). All sectors are calibrated to the base year 2010, for
which the existing stock of energy technologies and their characteristics are known and taken into account. A large variety of future
supply and demand technologies are represented by technoeconomic parameters such as capacity factor, energy efficiency,
economic lifetime, capital costs, O&M costs etc. The investment cost
assumptions for the most important electricity generation technologies are presented in Table A-2 in the Annex. The model also
includes assumptions for attributes not directly connected to individual technologies, such as energy prices, resource availability and
the potentials of renewable energy sources. UKTM has a temporal
resolution of 16 time-slices (four seasons and four intra-day timesslices). In addition to all energy flows, UKTM tracks CO2, CH4, N2O
and HFC emissions. For more information on UKTM see Ref. .
In addition to its academic use, UKTM is the central long-term
energy system pathway model used for policy analysis at the
Department of Energy and Climate Change (DECC) and the Committee on Climate Change (CCC).
2.2. Technology uncertainty dimensions
To arrive at a comprehensive picture of the potential impacts of
technology (and resource) uncertainty on the decarbonisation pathways in the UK, 5 key low-carbon technology dimensions have been
chosen for the sensitivity analysis: nuclear energy, carbon capture and
storage, bioenergy, renewable electricity and demand-side change.
For each dimension, a consistent narrative for, both, the central case
and the sensitivity variant, has been developed and then further
translated into quantitative model input assumptions (Table 1).
2.2.1. Nuclear energy (N)
According to the most recent government cost estimates, nuclear energy is currently the low-carbon technology with the
lowest generation costs in the UK  and is therefore at the centre
of the government’s decarbonisation strategy with a contribution
of up to 75 GW by 2050 (compared to the current 11 GW) according
to the UK’s Carbon Plan (scenario â€œHigher nuclear; less energy efficiencyâ€; .
But even though nuclear power constitutes a proven technology
and has contributed to electricity generation in the UK for more
than five decades, a number of uncertainties surround its future
development, most importantly with respect to costs and public
acceptance. Nuclear power costs have recently risen considerably,
leading to long delays in starting construction as well as difficulties
in finding investors. The future competitiveness of nuclear power
with other low-carbon technologies is far from certain [5,88].
Public acceptance of nuclear energy is generally relatively high in
the UK compared to other countries , but it remains to be seen
whether the possible delays and significant cost increases for the
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 155
first planned new-built plant since 1995 at Hinkley C could have a
Accordingly, in the central case, assuming that investments in
nuclear energy are rejuvenated and public opinion is favourable,
nuclear capacity is allowed to grow to a maximum of 33 GW until
2050 in line with the current central government estimate based on
the Dynamic Dispatch Model (DDM)  and the core scenario of
the Carbon Plan . In contrast, under the technology variant
nuclear (N), based on the aforementioned barriers in terms of
financial investors and public acceptance, we analyze the extreme
case of no new nuclear projects being realized in the UK until 2050.
2.2.2. Carbon capture and storage (C)
Carbon capture and storage (CCS) is the only technology with
which the continued use of fossil fuels in electricity generation and
industrial processes is compatible with a low-carbon transition.
CCS is therefore seen as a crucial component of a long-term emission reduction pathway in the UK. The important role of CCS
technologies had also been recognized by the UK government in
their CCS Roadmap from 2012 .
At the same time, CCS technology is at an early stage of development and has not been yet technically proven at full scale. The
important risk factors concerning the future deployment include
technology and constructions risk, the future competitiveness
compared to other low-carbon options, the financial feasibility of
CCS projects, infrastructure and storage risks, the regulatory
framework, as well as the public acceptance of CCS . The high
uncertainty is reflected in the current political situation in the UK
where the CCS Commercialisation Competition, providing Â£1 billion
capital funding, has just been cancelled .
Thus, in the central case of this sensitivity analysis, it is assumed
that these challenges are overcome and CCS technologies in electricity generation, industry and hydrogen production are commercially available from 2020 onwards (CCS plants with bioenergy from
2030). In addition, for all CCS technologies an annual capacity
growth constraint of 10% is applied. In light of the considerable risks,
the technology variant carbon capture and storage (C) assumes that
CCS technologies will unavailable in the UK before 2050.
2.2.3. Bioenergy (B)
Bioenergy can play a crucial role in decarbonisation strategies as
it is highly versatile. With usage options in electricity, heat and
transport, biomass offers dispatchable, predictable and controllable
energy output. Moreover, the potential of biomass with CCS to
deliver negative emissions is often highlighted as a crucial carbon
reduction technology . On the other hand, concerns are voiced
regarding the sustainability of biomass production for energy and
its impacts on food production and other environmental targets,
such as biodiversity . Additionally, the actual emission
reductions achievable with bioenergy may be considerably lower if
the lifecycle impacts are taken into account [46,47].
The bioenergy potentials and costs in this scenario analysis are
based on the Committee on Climate Change’s Bioenergy Review for
the UK . In our central case, we use the medium scenario
â€œExtended Land Useâ€ with total bioenergy potential reaching about
1300 PJ per year in 2030 compared to a biomass use of about 400 PJ
in 2015. The potential covers both imports and domestic resources
of dedicated energy crops, forestry and agricultural residues as well
as waste. In this scenario, less than 10% of the UK’s arable land
would be used for dedicated energy crops. This potential is then
held constant until 2050 assuming that both the maximum sustainable yield for domestic resources and the maximum import
volume are reached (in contrast to the CCC report where falling
bioenergy imports are expected). In the technology variant bioenergy (B), the projection for the domestic biomass resources is
based on the scenario â€œConstrained Land Useâ€, reflecting lower crop
yields and tighter social and environmental constraints on biomass
production where only 5% of the UK’s arable land are used for
dedicated energy crops. We also assume that no bioenergy imports
are available in the UK from 2020 onwards, limiting the available
biomass to around 380 PJ per year over the model horizon. In both
cases, biomass resources are assumed to be carbon-neutral
following the bioenergy emission accounting approach of the EU
Renewable Energy Directive . The optimization approach is free
to choose where and when the bioenergy is used, allowing us to
explore the competition between different sectors and technologies and the most cost-efficient deployment areas.
2.2.4. Renewable electricity (R)
The UK has considerable potential for renewable electricity generation,most importantly for onshore and offshore wind energy .
As of writing, the UK currently has the highest installed capacity of
offshore wind turbines in the world  and has experienced significant growth in solar PV installations . However, several issues
affecting the UK renewable industry need to be highlighted. Most
importantly, concerns are raised regarding the stability of government support [75,77], most clearly demonstrated in the recent deep
cuts to solar and onshore wind subsidies . The change in subsidy
level has strongly affected local supply chains. Wind power faces
considerable local opposition in the UK, mostly due to the visual
impacts [39,65]. Uncertainty also exist regarding the future cost
development of both less mature technologies, such as marine or
geothermal energy, and more mature technologies such as offshore
wind farms as they move to more challenging environments further
away from shore [40,41,70]. Lastly, the issue of integrating large
amounts of intermittent and decentralized renewable generation
into the UK power system needs to be recognised, leading to additional costs and possible constraints on their overall use .
Overview of technology dimensions.
Dimension Central case Sensitivity
Nuclear (N) New nuclear capacity limited to 33 GW until 2050 No additions after 2010
Carbon capture and storage
Capacity growth limited to 10% p.a.
Maximum capacity of 45 GW in electricity
Available in 2020 (2030 for bio-CCS)
CCS does not become available in the UK
Bioenergy (B) Total potential (imports Ã¾ domestic): 1300 PJ per
380 PJ per year
Renewable electricity (R) High technical potential (>400 GW)
Learning effects for all technologies
Restricted potential (49 GW)
Higher cost assumptions for offshore wind & solar PV
Marine & geothermal not available
Demand-side change (D) Medium elasticities (0.03 to 0.7)
Growth constraints of 10/15% p.a. on all
Low elasticities (0.01 to 0.5)
Growth constraints lowered to 5%/7.5% for innovative and energyefficient
156 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
In the scenario analysis, both the future potential and cost assumptions are varied. In the central case, a high technical resource
potential (>400 GW) and learning effects are implemented for
renewable sources in the UK, while under the technology variant
renewable electricity (R) the renewable potential is restricted to
51 GW (without hydropower) – reflecting acceptability as well as
system integration issues – and costs remain at the current level
(see Table A-3 in the Annex for more detail).
Some limitations need to be mentioned regarding the representation of intermittent renewables in UKTM. While the necessary
back-up capacity for intermittent renewables is accounted for, it has
to be noted that the spatial (UK as one region) and temporal granularity of our model (with 16 time-slices) is not ideal for capturing
the impacts of decentralized generation as well as the interplay
between varying demand and supply. Some system effects and costs
(in terms of required storage capacity, grid expansion and demand
response) are not fully reflected in a whole energy system model
such as UKTM requiring details operational simulation.
2.2.5. Demand-side change (D)
While the other technology dimensions focus primarily on supply technologies or resource availability, the dimension of demandside change (D) is introduced to acknowledge the importance of the
demand side for emission mitigation. While a variety of studies have
identified substantial abatement potentials at comparatively low or
even negative costs in the industry, buildings and transport sectors
(cf. for example [49,64], in reality many of these potentials are not
exploited and the rate of diffusion of novel technologies is considerably lower than what would be expected from the socially costoptimal perspective. This phenomenon, referred to as the energy
paradox or the energy-efficiency gap, can be explained by a variety of
market failures and barriers, such as limited information, transaction cost, risk adversity or market heterogeneity .
In energy system optimisation models, such barriers are incorporated by either limiting market share, applying hurdle rates or
constraining the diffusion rate of innovative technologies [28,67,94].
In this scenario analysis, the dimension D focuses on the diffusion rates of demand-side technologies and the demand response
to changing energy service prices. Historic diffusion rates in energy
systems have been analysed in several studies (cf. for example
[38,99]. Usually, growth rates between 10% and 15% are applied to
all new demand technologies in UKTM. However, given the evidence of diffusion rates of less than ten percent in some studies (cf.
for example , these rates are halved to 5%e7.5% for innovative
and highly efficient demand-side technologies under the technology variant D.
1 Moreover, the long-run own-price elasticities that
are attached to the different energy service demand categories, are
lowered to 0.01 to 0.5 instead of 0.03 and 0.7 in the central
case (based on Ref. .
2.3. Scenario matrix
The quantified descriptions of the five technology dimensions
(Table 1) are implemented in a comparative scenario analysis with
UKTM. To explore the uncertainty space of the availability of the
key low-carbon technology options in the UK, we conducted a
global sensitivity analysis with scenarios for all possible combinations of the five dimensions. With 5 dimensions, each with 2
possible values, a scenario matrix of 32 (25
) scenarios is established.
This allows us to assess both the individual effects of each dimension, and also to understand the interaction effects of the availability of several low-carbon options. In the results chapter, the
scenarios are denoted by the one-letter abbreviation of each
dimension that is restricted, e.g. the scenario where only nuclear
energy is restricted is labelled N, the scenario where all 5 dimensions are restricted is labelled NCBRD.
Apart from the technology assumptions described in Table 1, the
standard input parameters of UKTM are used in the scenario
analysis . The projections for energy service demands are based
on government forecasts of the development of household growth,
employment, transport demand etc. and are in line with average
annual growth rates of 2.1% for GDP and 0.4% for population (based
on the DECC EEP model2 and  (low migration variant)). The
assumptions for fossil fuel prices are taken from Ref. . With
respect to future technology costs, exogenous learning rates are
applied, especially in the case of less mature electricity and
hydrogen technologies, assuming that the UK is a price taker for
globally developing technologies. A global discount rate of 3.5% p.a.
for the first 30 years and 3% afterwards is used based on . In
addition, sector-specific discount rates are included to reflect the
varying private costs of capital by sector (10% for all energy supply
sectors as well as the industry, agriculture and service sectors, 7%
for transport and 5% for the residential sector; based on [12,81].
All scenarios take the UK legislation on GHG emission limits into
account, comprising the four five-yearly carbon budgets that have
been fixed so far until 2027  and the long-term target of an 80%
reduction until 2050 compared to 1990 . In order to give the
model flexibility with respect to the timing of emission reductions
after the already implemented carbon budgets, the long-term
target is applied via a cumulative budget constraint covering the
period from 2028 to 2050 which results in the same total quantity
of emissions as would a linear reduction pathway to 80% until 2050.
Thus, the model also has the option to comply with the cumulative
budget through early action, i.e. by realizing deeper emission cuts
in earlier periods allowing a higher emission level in later periods.
This means, that the cumulative target can be fulfilled without
actually reaching an 80% reduction until 2050 if in earlier periods
the emission reduction exceeds the linear reduction pathway.
3. Results of the uncertainty analysis
This section gives an overview on the most important outcomes
of the uncertainty analysis focusing on the variability between the
scenarios and the issue of complementarity or substitutability of
technologies. First of all, it needs to be pointed out that for three of
the most extreme cases, NCBR, NCRD and NCBRD, no feasible model
solution fulfilling the given constraints (most importantly the cumulative carbon constraint) can be generated such that these scenarios are excluded from the results analysis. This result should not,
however, be forgotten, as it suggests that with our assumptions the
UK’s carbon target can only be met if at least some of the technologies in our analysis perform according to expectations and
have the ability to diffuse strongly. The infeasible cases highlight
the circumstances under which the UK’s achievement of its climate
goals would be highly unlikely. If all other analysed technology
1 Lower diffusion rates are applied to:
Buildings: heat pumps, solar heating systems, micro-CHPs, conservation technologies; hydrogen pipelines to the residential and services sectors are disabled.
Transport: hydrogen and electric (including plug-in hybrids) vehicles.
Industry: all process technologies which are beyond the current state of the art;
hydrogen as well as high-efficiency boilers.
2 A description of the DECC EEP model can be found in Ref. . The actual model
runs underlying the demand projections for UKTM have not been published.
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 157
options fail, under a higher biomass potential or less restricted
demand-side diffusion, the model fails to reach the UK’s carbon
target under the given scenario assumptions.
3.1. The reference case
In order to set the reference point for the uncertainty analysis,
this section briefly describes the energy system results for the
reference case, in which all technologies follow the Central Case
assumptions (Fig. 1). The most significant changes occur in the
electricity system with a complete phase-out of electricity generation from coal by 2025. With respect to electricity generation, the
contribution of natural gas is also gradually reduced to almost zero
by 2050, while a capacity of 16 GW of gas-powered plants is still
required in 2050 as back-up capacity. In 2050, generation is strongly
dominated by nuclear power (66%) and bio-CCS (20%), whereas
wind power and other renewable source only play a minor role. Only
limited electrification of the end-use sectors can be observed in the
reference case with the total amount of electricity generated being at
almost the same level in 2050 as in 2015. Until 2025, there is even a
reduction in electricity demand due to efficiency gains (most
importantly for lighting), a reduction in the use of night-storage
heaters and falling demand levels in the industrial sector.
Given the significant decarbonisation efforts in the electricity
system, less drastic adjustments are required in the end-use sectors.
Final energy consumption is reduced by about 16% between 2015 and
2050 with the strongest reduction in the industrial sector. In terms of
fuel use, final energy demand is still dominated by fossil fuels with a
share of 62% in 2050 (compared to 81% in 2015). Due to the limited
availability of biomass and its strong use for bio-CCS in the supply
side, the role of bioenergy in the end-use sectors remains comparatively small. A gradual uptake of hydrogen, produced in centralized
CCS plants and covering about 11% of final energy demand in 2050,
occurs in car and light duty vehicle transport as well as in some
industrial sectors for low-temperature heat generation. Overall, this
relates to a reduction in primary energy consumption of 5% between
2015 and 2050. Total fossil fuel consumption is reduced by almost
40% in this period, but still covers more than half of primary energy
consumption in 2050. The substantial reduction before 2025 is predominantly due to the phase-out of inefficient coal power generation
and a declining energy demand in the industry sector (mostly
because of falling production levels).
As explained in Section 2.3, implementing the 80% emission
reduction target as a cumulative budget constraint gives the optimization approach more freedom with respect to the timing of the
mitigation efforts. In the reference case, total GHG emissions
decline by 76% by 2050 compared to 1990. The reference case
complies with the cumulative carbon budget constraint and favours
early action. Higher emission cuts are realised than a linear
reduction path would imply, especially between 2030 and 2040, so
avoiding investments in additional costly abatement options to
reach the ambitious level of 80% in 2050. As expected, emission
mitigation is dominated by the electricity sector which is carbonneutral by 2030 and accounts for 100 Mt of CO2eq of negative
emissions in 2050 (due to the use of carbon-neutral biomass with
CCS). With respect to the end-use sectors, the highest reductions
between 2015 and 2050 are achieved in the industry (59%) and
residential sector (49%) such that in 2050, the remaining GHG
emissions are dominated by the transport sector. The carbon price
(given in the model as the shadow price of the carbon constraint)
rises from Â£27/tCO2eq in 2020 to Â£244/t CO2eq3 in 2050.
It becomes obvious that in the reference case, decarbonisation is
highly dependent on large-scale technologies such as nuclear power plants and bio-CCS. This enables a delay to mitigation efforts in
2015 2030 2050
AGR & LULUCF
Electricity generaÆŸon Final energy consumpÆŸon
6170 PJ 5206 PJ
Coal Coal CCS Natural Gas Gas CCS Oil Biomass
Biomass CCS Wind Other RE Nuclear Hydrogen Electricity
Primary energy consumpÆŸon GHG emissions
Fig. 1. Overview on the reference case.
3 All monetary values stated in this paper are in real terms, with 2010 as base
158 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
the end-use sectors, most importantly in the transport sector. Yet,
from a policy perspective, this can be an unbalanced and especially
risky strategy as it depends on technologies and combination of
technologies and resources with highly uncertain future costs and
availability, which in some cases, have so far not even reached
3.2. Insights on variability across the whole scenario matrix
3.2.1. Variability in fuel use
Exploring the variability in fuel use across the entire scenario
matrix provides insights into the long-term uncertainty of decarbonisation pathways as a function of technology availability and
performance. The analysis focuses on the variability at the endpoint
of the model horizon, as represented in the box-and-whisker plots
for the most important fuel demands in 2050 shown in Fig. 1.
The strongest variability in 2050 can be observed for the use of
natural gas, both in terms of the interquartile range which spans
1500 PJ and in terms of the minimum (630 PJ) and the maximum
value (5500 PJ). Generally, restricting the various dimensions has a
dampening effect on the use of natural gas (i.e. the median value is
below the value in the reference case). As can be expected, the nonavailability of CCS has the strongest negative impact, whereas the
gas demand is increased in scenarios which feature CCS but have
restrictions on other low-carbon electricity options (i.e. scenarios
with R and N). However, while there are scenarios with increased
gas use in electricity generation, final use of gas is reduced in almost
all of the restricted cases, as many of the restrictions imply that
stronger mitigation efforts are required on the demand-side
compared to the reference case. Taking into account the uncertainty in technology development, this result suggests that the
pressure on demand side mitigation may be even stronger than
normally shown by scenarios exercises in which all technologies
perform as expected.
Given the relevance of oil products in the transport sector, there is
significantly less variability in oil use across the scenarios. Compared
to 2010, oil demand is reduced between 38% and 74% with road
transport fully covered by hydrogen and electricity in the more
restricted cases, but substantial shares of petroleum products still
required in aviation and shipping due to the limited availability of
biofuels. Use of oil is at its lowest when low-carbon electricity generation is strongly restricted (scenario NCR), such that the transport
sector has to contribute strongly to emission mitigation (through a
high use of hydrogen and a small contribution of biofuels).
While no clear trend towards stronger electrification can be
observed in the reference case, emission mitigation through a
higher use of electricity becomes more relevant in the restricted
scenarios, with a variety of factors and combination of factors
incentivising stronger electricity use. Both a limitation of the potential for negative emissions on the supply side (through restrictions on CCS and/or the bioenergy potential) and constraints
on the rate of demand-side change lead to a higher electrification of
the end-use sectors with the possibility of a doubling of electricity
demand compared to 2010 in some of the extreme cases. The most
significant increases in electricity use occur in the residential sector
through an increased use of heat pumps, which provide up to 61%
of residential heat demand in 2050. In some of the most restricted
scenarios (CBRD and NCBD), electric vehicles cover up to 80% of car
travel demand in 2050, but in general decarbonisation in the
transport sector relies much more strongly on hybrid vehicles and
hydrogen for heavy duty trucks. Less clear electrification trends can
be observed in the industry and services sector. Electricity prices
are not set exogenously but are determined within the model as the
marginal costs of electricity generation. Restricting low-carbon
generation options leads, as can be expected, to a significant
increase in these prices. However, in some of the more restricted
scenarios, electricity use will increase anyway as electrification
becomes one of the last options for emission mitigation.
The use of hydrogen, produced from biomass and/or in plants
equipped with CCS, constitutes an alternative decarbonisation
strategy for the demand side. However, the uncertainty analysis
shows that most of the restricted scenarios have a lower demand
for hydrogen in 2050 than in the reference case. The CCS dimension
has the strongest impact with hardly any hydrogen entering the UK
system in scenarios without CCS. Apart from that, the dimension of
demand-side change limits the uptake of hydrogen technologies,
especially in the transport sector. The highest contribution to final
energy demand (with a maximum of 18% in 2050) occurs in the
cases with restricted biomass potentials. Restricted availability of
biomass limits the deployment of bio-CCS in electricity generation
and puts more pressure on the end-use sectors to decarbonise.
To explore the use of renewable sources across the uncertainty
scenario matrix, biomass and other renewable energies are included
in the fuel-specific box-and-whisker plots in Fig. 2 and the variability
in the renewable share in gross final energy consumption, differentiated by electricity, heat and transport, is depicted in Fig. 3 (left-hand
side). Note that the shares in heat and transport only contain the
direct use of renewable resources in these sectors and not the use of
renewable electricity. To emphasise the importance of biomass in
decarbonisation strategies, is the results show how biomass is fully
exploited in all scenarios up to the maximum available in any scenario. However, the other restrictions placed on the other dimensions
have the effect of influencing where the biomass is used. For example,
while biomass is mainly used in electricity generation if CCS is
available, given the opportunity to generate negative emissions
through bio-CCS, an almost complete shift to the end-use sectors,
most importantly in the industry and commercial sector occurs in
scenarios without CCS. Bioenergy, if not further restricted, can then
cover up to 22% of total final energy demand in 2050.
While other renewables do not play a significant role in electricity
generation in the reference case, their contribution rises in almost all
of the restricted cases. Substantial amounts of on- and offshore wind
energy are deployed as soon as one of the other low-carbon electricity options is restricted, reaching an installed capacity of up to
105 GW in 2050 (in cases with high availability of renewable options).
Sizeable amounts of solar energy, with a maximum capacity of
45 GW, are only utilized in some of the more restricted cases (where
both nuclear energy and CCS are unavailable), while marine energy
only reaches a non-negligible role in the most restrictive cases for the
Fig. 2. Box-and-whisker plots on fuel use in 2050 across all 28 scenarios.4
4 All box-and-whisker plots shown in this paper depict the interquartile range in
the box, while the whiskers represent the maximum and minimum value.
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 159
electricity system (without limiting the renewable dimension, i.e.
NCB and NCBD). Accordingly, the renewable share in electricity
generation varies substantially between 18% and 97%in 2050, with an
intermittent share of up to 80%. A variety of studies have shown that
renewable shares of up to 80% would be technically feasible [14,71]
for the UK , for the European Union) e at a manageable cost.
The necessary back-up capacity for intermittent renewables reaches
up to 50 GW of natural gas or hydrogen fuelled gas turbines in some of
the scenario runs. Asmentioned above, other system effects and costs
of the increase in intermittent renewables are not fully reflected in a
whole energy system model like UKTM. In light of the strong
competition for bioenergy resources, biofuels for end-use do not
become a central decarbonisation strategy in any of the scenarios.
The renewable share in heating is strongly dominated by the use of
heat pumps in buildings, while high contributions from biomass are
only obtained in scenarios where bio-CCS is not available. On the
whole, the renewable share in gross final energy consumption varies
between 14% and 61% in 2050 and is at its lowest in scenario R.
Finally, we analyse the reduction in final energy consumption
across the scenario matrix as the last dimension on variability in fuel
use (right-hand side of Fig. 3). The reduction in final energy demand
by 2050 compared to 2010 lies between 13% and 37%. The strongest
reduction efforts occur in two types of scenarios: (1) in cases without
CCS, but strong potential for low-carbon electricity generation and
(2) in very restricted cases. While in the first group the residential
sector is the most relevant for the additional savings, the transport
sector plays a pivotal role in the very restricted scenarios. Constraining demand-side change only has a dampening effect on energy efficiency in the less restricted cases. Compared to the reference
case, the strongest additional savings are realized in the residential
and transport sector, whereas in the industrial sector the potential
for energy efficiency measures is already almost fully exploited
allowing further demand reductions only in a few extreme cases. In
contrast, based on the underlying growth in the sector’s GVA, a
considerable increase in energy demand is still expected in the
services sector, ranging between 17% and 33% in 2050. Interestingly,
the technology restrictions tend to raise the commercial fuel consumption, which is mainly driven by the lower demand elasticities
and restricted uptake of efficiency measures under variant D as well
as an increased use of biomass in scenarios without CCS.
From a policy perspective this sensitivity analysis suggests
important hedging strategies to achieve long-term deep decarbonisation. Even if the usual reference case with broad technology
availability proposes a strong focus on the electricity sector with
large-scale technologies like CCS and nuclear, the sensitivity analysis points towards a stronger use of renewable sources as well as
stronger efforts in terms of energy efficiency in the buildings and
transport sectors to ensure that long-term mitigation targets are
reached in the presence of technology uncertainty.
3.2.2. GHG emission reduction
With respect to emission mitigation we first investigate the
overall reduction trajectory. Implementing the 80% reduction target
for 2050 as a cumulative budget (between 2028 and 2050), offers
insights into the dependency of the optimal timing of the abatement actions on the technology availability. Fig. 4 shows the minimum and maximum GHG emission in each period as well as the
two most extreme pathways (in terms of emission reduction in
2050). There is a considerable spread in the reduction reached in
2050 from 67% (NCR) to 76% (REF) compared to 1990. Please note
that all scenarios with a feasible solution comply with the cumulative budget set in the model. However, none of the scenarios
fulfils the budget by reaching the politically set 80% reduction
target in 2050. Instead, early action is favoured such that especially
between 2030 and 2040 higher emission cuts are realised than a
linear reduction path would imply thereby avoiding investments in
additional, costly abatement options in 2050 (as reflected in the
difference between the hypothetical linear reduction path way
shown by the dotted line and the actual emission pathways shown
in Fig. 4). The tendency for early action clearly increases with the
number of technology restrictions in the scenarios. This highlights
that bringing forward ambitious mitigation efforts is a clear
hedging strategy against technology failure. Note that more scenarios would be infeasible if the climate target was implemented
by a fixed linear reduction pathway such that the model would
have to reach 80% by 2050. The still significant emission level in
the more restricted cases (around 250 Mt CO2eq in 2050) also
raises concerns regarding the ability of the system to arrive at netzero carbon emissions in the long term.
Strong variations can be observed in the sector-wise contribution
to emission reduction across the 28 scenarios (Fig. 5). While the
electricity sector reaches carbon-neutrality by 2035 in all scenarios,
the long-term ability to generate negative emissions through bioCCS strongly depends on the availability of CCS and biomass. In the
residential sector, mitigation efforts increase in almost all restricted
cases compared to the reference case, especially through the
exploitation of all available conservation measures and stronger
0% 20% 40% 60% 80% 100%
RE in electricity
RE in transport
RE in heaÆŸng
Renewable share in gross final
energy consumpÆŸon (2050)
-60% -40% -20% 0% 20% 40%
ReducÆŸon in final energy
consumpÆŸon (2050 to 2010)
Fig. 3. Box-and-whisker plots on the renewable share in gross final energy consumption (left) and reduction in final energy consumption (right) across all 28 scenarios.
2025 2030 2035 2040 2045 2050
Fig. 4. Range in GHG emission reduction pathways across all 28 scenarios (â€œLINEARâ€
shows the hypothetical linear reduction pathway to reach an actual 80% reduction in
160 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
electrification. The service sector exhibits the highest level of variation, evenwith the possibility of an increase in emissions compared
to 2010 in cases with limited demand-side change (including higher
demand levels) and electrification potential. In the industrial sector,
increased mitigation efforts occur especially in the cases with
limited biomass availability, but good potential for low-carbon
electricity generation. In contrast, particularly in the very
restricted cases without CCS, the industry sector’s contribution to
emission abatement is limited compared to the central case. The
transport sector is the only end-use sector that still makes a sizeable
contribution to GHG emissions in all scenarios in 2050. Contrary to
the other end-use sectors, transport sector abatement efforts increase in most of the very restricted cases compared to the reference
case through a stronger focus on efficiency measures.
3.2.3. Cost indicators
Apart from technology deployment and fuel use, the model
results also allow a comparison of the different scenarios in terms
of the implications for costs. Two types of cost metrics are analysed:
(1) total societal welfare costs, which are defined as the net total
surplus of producers and consumers and comprise the entire costs
of a specific energy system in a certain region and a certain period,
covering capital costs for energy conversion and transport technologies, fixed operating and maintenance costs as well as fuel and
certificate costs; (2) the carbon price, given in the model as the
shadow price of the budget constraint on GHG emissions.
In order to assess the importance of the single or combined
effect of the various technology dimensions, we examine the
change in cumulative welfare costs compared to the reference case
(Fig. 6). The cumulative cost increase is still relatively low (up to
2.4% compared to REF) in the scenarios where only one dimension
is restricted. While limiting renewable electricity options has little
impact, the strongest additional cost burden results from constraining the availability of CCS or biomass. Interestingly, given its
strong deployment in the reference case, removing nuclear energy
from the available technology options has no significant cost
impact since other low-carbon electricity options, most importantly wind energy, are available at similar cost levels.
Looking at the cases with two restrictions shows that the
combined cost effect of two dimensions is usually higher than the
sum of the two individual effects, with the exemption of scenario
CB, as the principal effect of both these dimensions is limiting the
deployment of bio-CCS plants. The strongest cost burden (up to
4.4% compared to REF) is caused by a combination of the dimensions C or B with a limited demand-side change (D) e putting
strong constraints both on the supply and demand side. Among the
scenarios with three technology restrictions, the highest impact on
welfare costs is induced either by severely limiting low-carbon
supply-side options (case NCR) or by a combined constraint on
crucial supply- and demand-side options (case CBD).
The extreme cases with four technology restrictions demonstrate which of the dimensions are crucial to keep the system-wide
cost impacts at a moderate level when all other analysed technology options are constrained. For the three cases with a feasible
solution, the increase in cumulative welfare cost, compared to the
reference case, is less than 7% in case NBRD and around 13% in case
NCBD revealing that both the availability of CCS and renewable
electricity options would keep the cost increase below 15%. In
contrast, costs increase a lot more in the case with nuclear energy
(CBRD), while reaching the long-term emissions target is not even
feasible with a higher biomass potential or less restricted demandside change when all other large-scale supply-side options fail. This
emphasizes the importance of the electricity sector in decarbonisation efforts as the long-term target can only be complied with if
at least one of the central low-carbon electricity options is available
at a significant level.
Similar to societal welfare costs, carbon prices vary considerably
across the scenario matrix with values in 2050 of Â£244/t CO2eq in
the central case and Â£7000/t CO2eq in scenario NRC (see Table A-4
in the Annex). Regarding the scenarios with 4 technology restrictions, the carbon price in scenarios NBRD and NCBD is significantly lower than in NRC (with Â£1500/tCO2eq and Â£3100/tCO2eq),
highlighting once more the importance of having either CCS or
renewable electricity options available to limit the decarbonisation
costs. The scenario CBRD is a clear outlier, reaching a carbon price of
almost Â£38,000/tCO2eq in 2050. This also shows that the carbon
price, being at the margin, is much more sensitive to technology
changes than the total welfare is. Adding, for example, a biomass or
CCS restriction on top of any combination of other restrictions
(including none e the central case) always at least doubles the
carbon price. Please note that the carbon price would be
Fig. 5. GHG emission reduction (left) and share in total emissions in 2050 (right) by sector across all 28 scenarios.
Number of restricted dimensions
1 2 3 4
Change in cumulated energy system
costs compared to REF
Outlier: CBRD, 33%
Fig. 6. Change in total welfare costs (discounted, cumulated from 2010 to 2050)
compared to the reference case.
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 161
significantly higher, if the carbon target was implemented by
enforcing an actual reduction of 80% until 2050 (and a linear
reduction pathway) instead of the cumulative budget approach
used in this analysis. The results of the carbon price also raise
concerns regarding the political feasibility of reaching the longterm climate goals in a technology-constrained world. Even if all
of the scenarios shown in Fig. 6 complying with the cumulative
carbon budget were technically feasible, the resulting carbon prices
would in most likelihood not be politically enforceable cf. also .
The ranking of scenarios according to welfare cost or carbon
price are generally quite similar. A few noteworthy exemptions
provide some indications with respect to the shape of the abatement cost curve. For example, the scenarios CR and CBR rank much
worse in terms of carbon price than welfare costs indicating a steep
abatement cost curve, while for the scenarios BD and NBD it’s the
We finish our discussion on costs with regression results for the
cumulated welfare costs, with the dimensions and their interactions
as independent variables.5 The coefficients in Fig. 7 give an estimate
of how much the existence of given scenario dimensions, or the
interactions between dimensions, adds to the cost metric across the
full scenario set. CBD, for example, would refer to the specific interactions effects between drivers C, B and D, excluding the effects
these drivers have by themselves and the interaction effects the
pairs of these drivers have (i.e. CB, CD, BD). Thus, the numbers in
Fig. 7 can be interpreted as follows: for one dimension they show the
increase in costs that occurs when this restriction alone is added to
the scenarios (e.g. welfare costs increase by about Â£20 billion when
nuclear energy is restricted), for several dimensions they show the
cost increase caused by the interaction effect (on top of the effect of
the individual dimensions; hence, welfare costs also increase by
about Â£40 billion due to the interaction of N and C, on top of the cost
increases caused by N and C individually).
While the results confirm the general trends discussed above, it
also highlights better the role of the interactions between the
drivers. For example, scenarios CBD and CBR have comparable costs,
but the regression analysis highlights how the welfare costs of the
latter are especially driven by the interaction between the three
drivers, whereas for the former, the impact of the individual drivers
and the interactions between two driver elements explains most of
the cost. It also shows that some dimensions only have a significant
impact when combined with other restrictions. For example,
restricting renewables alone, or in combination with only one other
dimension, does not have a strong impact on welfare costs, whereas
the interaction effects of dimension R with two other dimensions
tend to be particularly high. Thus, restricting the development of
renewable electricity generationwill drive the costs of the UK energy
transition significantly if other options, like nuclear energy or CCS,
turn out to fail. This emphasizes again the importance of a balanced
portfolio of low-carbon technology options.
3.3. Insights on technology use
3.3.1. Complementarity and substitutability of technologies
To gather further insights into the impact of uncertainty in the
availability of low-carbon options on the long-term development of
the UK energy system, we now look at technology complementarity
and substitutability. To do this we assess which technologies are
generally only used in combination with one another and which
can easily substitute for each other. We calculate the correlation
coefficients between the 5 technology dimensions (with a value of
zero if the technology is restricted and of 1 if not) and the use of the
most important fuels in 2050 (total or by sector). In Table 2 all
coefficients with an absolute value over 0.5 are highlighted which
are further explored in the following text.
As already pointed out in Section 3.2, the use of natural gas is
highly variable across the scenario matrix. A strong positive correlation of gas use, especially in hydrogen production, with CCS
availability can be observed, highlighting the fact that significant gas
use is only possible, in the long-term, if CCS is available. In contrast,
less gas is needed if nuclear energy and renewable electricity options
succeed in the long-term. Finally, gas demand is not strongly
correlated with the resource potential of bioenergy. In the residential sector (and to a lesser extent in the industrial sector), there is a
strong positive correlation between dimension B and gas use, indicating that if higher amounts of bioenergy are available, there is less
pressure to reduce gas use in heating in the end-use sectors.
Further investigation into the trend towards electrification reveals that total electricity demand in 2050 is only significantly
correlated with biomass availability, i.e. if the bioenergy potential is
strongly limited, greater electrification is required, especially in the
industry and residential sector, while no such trend is observable
when restricting the other dimensions. At the sector level, a more
considerable uptake of electric heating and other devices takes
place in the services and residential sectors if demand-side change
is constrained, balancing mostly the higher overall energy demand
and the slower realization of conservation measures. In transport,
electricity use is negatively correlated with CCS availability, highlighting that in the scenario analysis at hand, the long-term
decarbonisation of the transport sector relies mainly on
hydrogen, while a substantial uptake of electric vehicles only occurs if the option to produce hydrogen in CCS plants does not exist.
The negative correlation between electricity use in hydrogen production and dimension C is of little relevance, given the generally
low level of hydrogen produced from electricity.
Amongst the highest correlations between technology dimensions and fuel use in 2050 is the positive one between
hydrogen use and CCS, emphasizing that CCS is essential to enable a
strong contribution of hydrogen to the UK decarbonisation
pathway. While in electricity generation a broader portfolio of lowcarbon options is available, hydrogen generation strongly depends
on CCS or bioenergy, where strong competition with other sectors
needs to be taken into account.
Regarding the use of biomass, it has already been mentioned
that the full potential is always exploited independent of the scenario assumptions. Accordingly, all correlations of total biomass use
and the other technology dimensions are close to zero. In contrast,
on the sector level, bioenergy demand is significantly correlated
with CCS availability. Hence, a substantial amount of biomass is
only used in electricity generation if the option of generating
negative emissions through bio-CCS is given. In all scenarios
without CCS, biomass in the end-use sectors is increased considerably, explaining the strong negative correlations, especially in the
industry sector. However, this is not the case for the transport
sectors, where biofuels never play a substantial long-term role.
In addition to the significant correlations highlighted in Table 2,
it is also interesting to examine some of the â€œnon-correlationsâ€
which might be rather unexpected. First of all, the use of CCS is not
correlated with biomass availability (correlation between dimension B and sequestered CO2 emissions). Thus, even if the possibility
5 The design of a simple OLS-regression is used here with WC Â¼ a Ã¾ b1*N Ã¾ â€¦ Ã¾
bn*D Ã¾ g1*NC Ã¾ â€¦ Ã¾ gn*RD Ã¾d1*BRD Ã¾ â€¦ Ã¾ dn*NRD Ã¾ q1*NBRD Ã¾ q2*NCBD; with
WC Â¼ cumulated welfare costs; the independent variables represent the 5 dimensions and their interactions and have a value of 1 if the dimension is restricted
and of 0 otherwise. In addition to the infeasible scenarios, the scenario CBRD is
excluded as it is an outlier and would strongly bias the results. As the difference in
technology dimensions is the only thing that is varied between the scenarios, the
regression has an R2 value of one, i.e. the difference in technology dimensions and
their interaction effects can perfectly describe the cost of the system.
162 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
for bio-CCS is limited, the total use of CCS is not affected. Secondly,
the insignificant correlation between the level of electrification and
CCS availability indicates that while a limited bioenergy potential
drives a higher electricity use, this is not the case for CCS. Lastly,
there is a relatively low correlation between the available biomass
resources and the use of bioenergy in industry showing that the
industrial biomass demand is comparatively robust against changes
in the biomass availability.
3.3.2. Technology diffusion
Another interesting dimension of future technology use are the
diffusion rates of key decarbonisation technologies showing how
ambitious some of the trajectories would have to be. Here, a special
look is taken at some of the end-use sectors which have to deliver a
significant contribution to emission abatement, especially in the
case of multiple technology failure on the supply side.
In car transport (Fig. 8), most scenarios feature high diffusion
rates for hybrid vehicles (including plug-in hybrids) reaching a
market share of around 40% (at the median) in 2030. In most scenarios, their market share starts to decline after 2040 (from up to
90%), as more ambitious decarbonisation technologies diffuse more
strongly. The share of electric cars remains below 20% in most
scenarios, with some considerable exceptions of contributions of
up to 30% in 2030 and 80% in 2050 (e.g. in scenario CBRD or NCBD).
Hydrogen vehicles have, on average, a slightly higher market share
then electric cars, but only reach a maximum of around 50% in 2050
(e.g. in scenario NBRD). The very wide range across the technologies suggest that alternative transport futures can be imagined and
these are likely to depend not only on the success of the transport
technologies themselves, but also on the success of the technologies that are further up the supply chain (e.g. H2 production, CCS,
low carbon electricity).
In the residential sector (Fig. 9), the abatement strategy relies
strongly on heat pumps which exhibit highly ambitious diffusion
rates reaching the maximally allowed share in the model of 60%
already by 2040. This is mirrored by a gradual decline in the use of
gas boilers, where the median share drops from over 80% in 2010 to
Fig. 7. Coefficients of an OLS-regression on cumulated welfare costs.
Correlation coefficients between technology dimensions and fuel use in 2050
(coefficients 0.5 are highlighted).
NC B RD
Gas, total Â¡0.50 0.73 0.12 Â¡0.50 0.05
Gas, ELC 0.44 0.45 0.27 0.37 0.20
Gas, HYG 0.27 0.66 0.29 0.26 0.25
Gas, residential 0.00 0.24 0.61 0.08 0.11
Gas, services 0.10 0.35 0.12 0.15 0.01
Gas, industry 0.24 0.22 0.44 0.22 0.01
Gas, transport 0.03 0.32 0.45 0.19 0.30
Electricity, total 0.18 0.12 Â¡0.57 0.26 0.40
Electricity, residential 0.11 0.06 Â¡0.51 0.24 Â¡0.50
Electricity, services 0.12 0.41 0.36 0.15 Â¡0.59
Electricity, industry 0.18 0.14 Â¡0.53 0.16 0.28
Electricity, transport 0.08 Â¡0.90 0.27 0.22 0.13
Electricity, HYG 0.31 Â¡0.70 0.13 0.25 0.15
Hydrogen, total 0.27 0.66 0.28 0.26 0.24
Hydrogen, end use 0.13 0.78 0.18 0.11 0.43
Renewables, total 0.31 0.46 0.31 0.75 0.13
Renewables, ELC 0.31 0.46 0.31 0.75 0.14
Biomass, total 0.01 0.01 0.92 0.01 0.12
Biomass, ELC 0.10 0.64 0.67 0.13 0.03
Biomass, end-use 0.04 Â¡0.75 0.60 0.06 0.11
Biomass, residential 0.07 Â¡0.61 0.48 0.05 0.31
Biomass, services 0.03 Â¡0.76 0.46 0.02 0.11
Biomass, industry 0.11 Â¡0.95 0.20 0.15 0.14
Biomass, transport 0.03 0.21 0.95 0.03 0.03
CCS (sequestered emissions) 0.31 0.91 0.02 0.28 0.04
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 163
22% in 2050. However, in some of the less constrained scenarios,
gas boilers still contribute up to 58% to residential heating in 2050.
The share of hydrogen-based heating systems (both boilers and fuel
cell micro-CHPs) in the residential sector remains negligible in
most scenarios. Apart from that, moderate contributions of standalone electric heating systems and night storage heaters can be
observed in most scenarios. The share of district heating in the
residential sector remains below 10%.
This analysis has shown how uncertainties in key low-carbon
options, covering nuclear energy, CCS, bioenergy, renewable electricity technologies as well as demand-side change, can influence
long-term decarbonisation pathways of the energy system and how
multiple technology failures can interact to produce impacts that
are more than additive. Using a technology-oriented, comprehensive energy system model allowed us to cover all the relevant repercussions within the energy system and to assess trade-offs
between sectors and mitigation efforts. The approach of a global
sensitivity analysis, applied here to a case study of the UK, can be
easily transferred to other national or supranational settings.
By systematically varying the availability of one or several
crucial mitigation options, critical insights can be gained on:
Variability across the uncertainty space. The analysis has identified an enormous range in the demand for natural gas across the
scenario matrix, while other fuel uses are less variable. Also, the
sector-wise contribution to emission abatement is highly sensitive to technology failure.
Complementarity and substitutability of technologies. Some
technology options only play a significant role in the decarbonisation pathway if other technologies are also available,
seen, for example, in the strong dependency of hydrogen technologies on the prevalence of CCS. At the same time, some
technologies can easily be substituted by others without
entailing a significant cost increase. E.g., the analysis at hand
indicates that in the UK energy system, replacing nuclear energy
with other low-carbon electricity options would not cause a
significant rise in electricity generation cost. From a risk
perspective, the reliance on bio-CCS in some scenarios is interesting as it combines the unproved CCS with a resource with
Critical low-carbon options and hedging strategies. The use of some
mitigation options is comparatively robust across the sensitivity
analysis, e.g. the early decarbonisation of the electricity sector or
considerable energy efficiency measures in the buildings and
industry sector. The evaluation of cost parameters also reveals
which options are most relevant to avoid a prohibitive cost
burden in the case of multiple technology failure (especially CCS
and renewable electricity options in this analysis). The sensitivity
approach can also help to identify â€œfailedâ€ low-carbon technologies which never play a substantial role, such as ocean (wave
and tidal) technologies in the present case. Moreover, the
sensitivity analysis can reveal crucial hedging strategies in the
presence of technology uncertainty that the reference case alone
might not show, like a higher use of renewable sources or
stronger focus on energy efficiency. For some of these options,
such as efficiency measures in the buildings sector, policy measures will have to overcome significant investment barriers, but
the technologies involved are a lot more mature and less risky
than some of the options chosen in the reference case.
Timing and path dependencies. While the analysis at hand has
mostly focused on the endpoint of the model horizon, i.e. 2050,
the sensitivity approach can also be used to assess the timing
and pathways of the decarbonisation challenge. The scenario
analysis clearly highlights that multiple technology uncertainty
strongly increases the importance of early action in emission
mitigation. Furthermore, analysing the diffusion rates of critical
technologies indicates the ambition of low-carbon pathways
In terms of future research, there are several possibilities to
advance work sensitivity analysis presented here. First of all, the
2015 2020 2025 2030 2035 2040 2045 2050
2015 2020 2025 2030 2035 2040 2045 2050
2015 2020 2025 2030 2035 2040 2045 2050
Hybrid vehicles Electric vehicles Hydrogen vehicles
Share in total car
Fig. 8. Diffusion rates in car transport across the scenario matrix (MED Â¼ median, grey area indicates the interquartile range).
Fig. 9. Diffusion rates in residential heating across the scenario matrix (MED Â¼ median, grey area indicates the interquartile range).
164 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
approach could be extended to account for further uncertainties.
Here, integrative techniques, e.g. by including policy makers and
other stakeholders or using expert elicitation , to identify the
key uncertainties in the energy system should be further developed
and tested. Secondly, while this analysis focuses on strong narratives and â€œin/outâ€ options of technologies like CCS or nuclear, the
economic and performance parameters of such technologies could
also be varied over sensible ranges to explore robust ranges and
tipping points in technology use. Thirdly, in order to examine the
uncertainty space further, alternative uncertainty approaches
should be applied to national energy systems analyses, e.g. stochastic methods or an assessment of the near-optimal solution
space through MGA.
For policy makers, a deeper understanding of the main uncertainties in mitigation strategies as well as their interdependencies is crucial to formulate more robust
decarbonisation strategies. The analysis has shown how multiple
technology failures could put the achievement of long-term emission reduction targets in the UK at considerable risk. It has also
shown how the failure or success, of a given technology can have
wider impacts elsewhere in the energy system, thus complicating
the management of a long-term energy transition. Further investigation is needed regarding the appropriate policy response to such
an uncertain environment, i.e. should the government support a
broad technology portfolio or move to â€œpicking winnersâ€ early on?
Sensitivity analysis, such as the one at hand, help to identify the key
low-carbon technologies and resources to which the low-carbon
transition is particularly vulnerable and to raise the awareness in
policy making to the extent of the uncertainty challenge.
This research was supported under the Whole Systems Energy
Modelling Consortium (WholeSEM) e Ref: EP/K039326/1.
Background on modelling techniques to represent uncertainty in energy-economic models.
Sensitivity analysis The most commonly used method to evaluate uncertainty in deterministic models analysing the variability of the model output as a
function of changing input parameters. In most cases, local or one-at-a-time sensitivity analyses are conducted, which, however, do not
capture the interactions between input parameters . This has led to a couple of studies using global sensitivity techniques, which
vary several uncertain input parameters at a time to explore the interaction effects, in some cases through probabilistic and Monte Carlo
methods . Recent studies in energy systems research include [1,2,6,73,92] and 
Stochastic modelling Stochastic modelling moves away from the deterministic approach still applied in sensitivity analysis and deals explicitly with optimal
decision-making under uncertainty by applying probabilities to unknown future parameters [3,61]. This has the advantage of
accounting for the cost of uncertainty, relaxing the assumption of perfect foresight and analysing hedging strategies. Stochastic
programming has been applied to a large variety of input assumptions in energy systems modelling, like energy prices , resource
availability , technology parameters , climate sensitivity , the stringency of mitigation targets  and the stochasticity of
intermittent renewable resources .
Modelling to Generate
An optimization technique that explores the near-optimal solution space for feasible solutions that are maximally different from the
optimal pathway . This technique takes into account that the single solution of an optimization approach never reflects the full
uncertainty and practical constraints of the real-world system and allows to assess both common features and significant differences in
the near-optimal solution space. Originally mostly applied in land and water management (e.g. Ref. , first applications in long-term
energy optimization models can be found [18,86,93].
Multi-model comparisons Multi-model comparisons are increasingly applied to examine both parameter and structural (or model) uncertainty. Such studies
involve running a predefined set of scenarios in several modelling frameworks (mostly Integrated Assessment Models) with diverse
model structures and input parameters (with different degrees of harmonization) . The aim is to arrive at a range of plausible
mitigation pathways, to explore the impact of different model structures and to assess the relevance of different input parameters.
Multi-model comparison studies have been pioneered by the Energy Modelling Forum (EMF) ; [55,57], but have since also been
carried out through projects like AMPERE , LIMITS  and the Asian Modeling Exercise (AME) . A recent review can be found in
Investment cost assumptions for electricity generation technologies in UKTM (selection) [Â£/kW].
2020 2030 2040 2050
Coal CCS 2707 2707 2700 2700
Gas, OCGT 287 287 286 286
Gas, CCGT 561 561 561 561
Gas, CCGT CCS 1185 1185 1179 1179
Nuclear 4019 4019 3973 3973
Biomass combustion, large 2324 2280 2275 2275
Biomass combustion, CCS 3626 3626 3610 3610
Biogas, gas engine 3885 3791 3780 3780
Central case 1477 1408 1398 1398
Restricted 1477 1408 1398 1398
Central case 2229 1982 1764 1764
Restricted 2472 2472 2472 2472
Central case 786 523 523 523
Restricted 980 980 980 980
Central case 1499 1190 1190 1190
Restricted 1713 1713 1713 1713
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 165
 B. Anderson, E. Borgonovo, M. Galeotti, R. Roson, Uncertainty in climate
change modeling: can global sensitivity analysis Be of help? Risk Anal. 34
(2014) 271e293, http://dx.doi.org/10.1111/risa.12117.
 D. Anthoff, R. Tol, The uncertainty about the social cost of carbon: a decomposition analysis using fund, Clim. Change 117 (3) (2013) 515e530, http://
 E. Beale, On minimizing a convex function subject to linear inequalities, J. R.
Stat. Soc. B 17 (1955) 173e184.
 BEIS, Solar Photovoltaics Deployment, Department for Business, Energy &
Industrial Strategy, 2016. Retrieved from the GOV.UK website, https://www.
 Birmingham Policy Commission, The Future of Nuclear Energy in the UK,
University of Birmingham, 2012.
 F. Branger, L. Giraudet, C. Guivarch, P. Quirion, Global sensitivity analysis of an
energyeeconomy model of the residential building sector, Environ. Model.
Softw. 70 (2015) 45e54, http://dx.doi.org/10.1016/j.envsoft.2015.03.021.
 E.D. Brill Jr., J.M. Flach, L.D. Hopkins, S. Ranjithan, MGA: a decision support
system for complex, incompletely defined problems, IEEE Trans. Syst. Man
Cybern. 20 (4) (1990) 745e757.
 K. Calvin, L. Clarke, V. Krey, G. Blanford, K. Jiang, M. Kainuma, E. Kriegler,
G. Luderer, P.R. Shukla, The role of Asia in mitigating climate change: results
from the Asia modeling exercise, Energy Econ. 34 (Supp 3) (2012) S251eS260,
 Carbon Trust, Accelerating Marine Energy, The Potential for Cost Reduction e
Insights from the Carbon Trust Marine Energy Accelerator. London, UK, 2011.
 CCC, Time Preference, Costs of Capital and Hidden Costs: a Committee on
Climate Change Note, 2011. Retrieved from the CCC website, http://archive.
 CCC, Bioenergy Review, Committee on Climate Change, London, 2011.
 CCC, The Renewable Energy Review, Committee on Climate Change., London,
 CCC, Fourth Carbon Budget Review e Part 2, The Cost-effective Path to the
2050 Target, Committee on Climate Change, London, 2013.
 S.-Y. Chang, E.D. Brill Jr., L.D. Hopkins, Efficient random generation of feasible
alternatives: a land use example, J. Regional Sci. 22 (3) (1982) 303e314.
 H.E. Daly, K. Scott, N. Strachan, J. Barrett, Indirect CO2 emission implications of
energy system pathways: linking IO and TIMES models for the UK, Environ.
Sci. Technol. 49 (17) (2015) 10701e10709, http://dx.doi.org/10.1021/
 J. DeCarolis, Using modeling to generate alternatives (MGA) to expand our
thinking on energy futures, Energy Econ. 33 (2) (2011) 145e152, http://
 DECC, The Carbon Plan: Delivering Our Low Carbon Future, Department of
Energy and Climate Change, London, 2011.
 DECC, UK Renewable Energy Roadmap, Department of Energy and Climate
Change, London, 2011.
 DECC, DECC Dynamic Dispatch Model (DDM), Department of Energy and
Climate Change, London, 2012.
 DECC, CCS Roadmap – Supporting Deployment of Carbon Capture and Storage
in the UK, Department of Energy and Climate Change, London, 2012.
 DECC, Electricity Generation Costs (December 2013), Department of Energy
and Climate Change, London, 2013.
 DECC, DECC and Industry Sets Out Plan to Stay â€œworld Leaderâ€ in Offshore
Wind and Boost Economy, 2014, February 19. Retrieved from https://www.
 DECC, Updated Energy and Emissions Projections 2014, 2014. Retrieved from
the GOV.UK website, https://www.gov.uk/government/uploads/system/
 DECC, Fossil fuel Price Projections: 2014, 2014. Retrieved from the GOV.UK
 DECC, Changes to Renewables Subsidies, 2015, December 17 (press release).
Retrieved from https://www.gov.uk/government/organisations/departmentof-energy-climate-change.
 P.E. Dodds, W. McDowall, Methodologies for representing the road transport
sector in energy system models, Int. J. Hydrogen Energy 39 (5) (2014)
 O. Edenhofer, K. Lessmann, C. Kemfert, M. Grubb, J. Kohler, Induced technological change: exploring its implications for the economics of atmospheric
stabilization: synthesis report from the innovation modeling comparison
project, Energy J. Endogenous Technol. Change (Special Issue #1) (2006)
 P. Ekins, G. Anandarajah, N. Strachan, Towards a low-carbon economy: scenarios and policies for the UK, Clim. Policy 11 (2) (2011) 865e882, http://
 Enviros, The Cost of Supplying Renewable Energy, A report for DTI, Enviros
Consulting, London, 2005.
 European Climate Foundation, Roadmap 2050: a Practical Guide to a Prosperous, Low-carbon Europe, 2010. Retrieved from the roadmap2050 website,
Assumptions under technology dimension renewable electricity (R).
Source Central case Sensitivity renewable electricity (R)
Potential: 36.9 GW 
Capital cost: Current DDM assumptions , 12% between 2010 and 2050 (to 1400
Potential: 15 GW 
Capital cost: same as in central
Potential: 327 GW ; limited to 49 GW until 2030 (amount leased so far)
Capital cost: -35% between 2010 and 2050 (to 1800 Â£/kW) 
Potential: 16 GW (current total with planning permission 
Capital cost: no learning effects after 2015, costs at 2500 Â£/kW in
Solar PV Potential: 45 GW (current DDM assumptions)
Capital cost: Current DDM assumptions, 47%/34% between 2010 and 2050 (to 520/
1190 Â£/kW (farm/rooftop))
Potential: 20 GW 
Capital cost: no learning effects after 2015, costs at 980/1700
Marine Potential: 27 GW 
Capital cost: Current DDM assumptions, with 7500 Â£/kW (wave) & 3900 Â£/kW (tidal
stream) in 2050
Geothermal Potential: 2 GW (current DDM assumptions)
Capital cost: Current DDM assumptions, with 3900 Â£/kW in 2050
Development of the carbon price [Â£/t CO2eq] in the 28 scenarios.
2020 2030 2040 2050 Increase in 2050
compared to REF
REF 27 128 180 244
N 24 139 196 266 9%
C 26 320 450 612 150%
B 27 273 384 522 114%
R 25 128 180 244 0%
D 25 190 267 363 49%
NC 26 383 540 734 200%
NB 25 295 415 564 131%
NR 24 144 202 275 13%
ND 24 204 287 390 60%
CB 25 453 639 868 255%
CR 26 393 554 752 208%
CD 25 400 564 767 214%
BR 26 287 405 550 125%
BD 24 380 535 727 198%
RD 27 192 271 368 50%
NCB 25 890 1255 1705 598%
NCR 31 3689 5200 7063 2791%
NCD 26 621 875 1189 387%
NBR 24 356 502 681 179%
NBD 25 438 618 840 244%
NRD 24 213 301 408 67%
CBR 35 1819 2565 3484 1326%
CBD 24 886 1249 1696 594%
CRD 28 816 1150 1562 539%
BRD 24 455 641 871 257%
NCBD 25 1637 2307 3134 1183%
NBRD 25 775 1092 1484 507%
CBRD 33 19668 27724 37661 15313%
166 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
 Energiewirtschaftliches Institut an der Universitâ‚¬
at zu Koln (EWI), Gesellschaft â‚¬
fÃ¼r Wirtschaftliche Strukturforschung (GWS), Prognos AG, Energieszenarien
fÃ¼r ein Energiekonzept der Bundesregierung, 2010 (Report for Federal Ministry of Economy and Technology. Basel, Koln, OsnabrÃ¼ck) â‚¬ .
 European Commission (EC), Directive 2009/28/EC of the European Parliament
and of the Council of 23 April 2009 on the Promotion of the Use of Energy
from Renewable Sources and Amending and Subsequently Repealing Directives 2001/77/EC and 2003/30/EC, European Commission (EC), Brussels,
 B. Fais, N. Sabio, N. Strachan, The critical role of the industrial sector in
reaching long-term emission reduction, energy efficiency and renewable
targets, Appl. Energy 162 (2016) 699e712, http://dx.doi.org/10.1016/
 Fais, B., Daly, H. E., & Dodds, P. E. (). UK TIMES Model Documentation, Version
1.0. London. UCL Energy Institute. (Forthcoming).
 A. GrÃ¼bler, N. Nakicenovic, D. Victor, Dynamics of energy technologies and
global change, Energy Policy 27 (5) (1999) 247e280, http://dx.doi.org/
 C. Haggett, Understanding public responses to offshore wind power, Energy
Policy, Special Sect. Offshore wind power Plan. Econ. Environ. 39 (2011)
 P. Heptonstall, R. Gross, P. Greenacre, T. Cockerill, The cost of offshore wind:
understand-ing the past and projecting the future, Energy Policy, Model.
Transp. (Energy) Demand Policies 41 (2012) 815e821, http://dx.doi.org/
 P. Higgins, A. Foley, The evolution of offshore wind power in the United
Kingdom, Renewa-ble Sustain. Energy Rev. 37 (2014) 599e612, http://
 HM Government, Climate Change Act 2008, HM Government, London, 2008.
 HM Government, HM Government Statement to Markets Regarding Carbon
Capture and Storage Competition, HM Government, London, 2015, November
 HM Treasury, The Green Book, HM Treasury, London, 2011.
 M. Hoâ‚¬ok, J. Li, K. Johansson, S. Snowden, Growth rates of global energy sys- â‚¬
tems and future outlooks, Nat. Resour. Res. 21 (1) (2012) 23e41, http://
 International Energy Agency (IEA), Technology Roadmap – Biofuels for
Transport, Organisation for Economic Co-operation and Development (OECD)/
International Energy Agency (IEA), Paris, 2011.
 International Energy Agency (IEA), Technology Roadmap – Bioenergy for Heat
and Power, Organisation for Economic Co-operation and Development
(OECD)/International Energy Agency (IEA), Paris, 2012.
 International Energy Agency (IEA), World Energy Outlook 2015, Organisation
for Economic Co-operation and Development (OECD)/International Energy
Agency (IEA), Paris, 2015.
 International Energy Agency (IEA), Energy Technology Perspectives 2015 e
Mobilising Innovation to Accelerate Climate Action, Organisation for Economic Co-operation and Development (OECD)/International Energy Agency
(IEA), Paris, 2015.
 IPCC, Intergovernmental Panel on Climate Change Fifth Assessment Report
(AR5), IntCergovernmental Panel on Climate Change, Geneva, Switzerland,
 A.B. Jaffe, R.N. Stavins, The energy-efficiency gap – what does it mean? Energy
Policy 22 (10) (1994) 804e810, http://dx.doi.org/10.1016/0301-4215(94)
 R. Kannan, N. Strachan, S. Pye, G. Anandarajah, N. Balta-Ozkan, UK MARKAL
Model Documentation, 2007. Retrieved from the UCL Energy Institute Models
 A. Kanudia, R. Loulou, Robust responses to climate change via stochastic
MARKAL: the case of Quebec, Eur. J. Operational Res. 106 (1) (1998) 15 e30,
 I. Keppo, B. van der Zwaan, The impact of uncertainty in climate targets and
CO2 storage availability on long-term emissions abatement, Environ. Model.
Assess. 17 (1e2) (2012) 177e191, http://dx.doi.org/10.1007/s10666-011-
 B. Knopf, Y.-H.H. Chen, E. De Cian, H. Forster, A. Kanudia, I. Karkatsouli, â‚¬
I. Keppo, Beyond 2020-Strategies and costs for transforming the European
energy system, Clim. Change Econ. 4 (Suppl.) (2013), http://dx.doi.org/
 V. Krey, D. Martinsen, H. Wagner, Effects of stochastic energy prices on longterm energy-economic scenarios, Energy 32 (12) (2007) 2340e2349, http://
 E. Kriegler, J.P. Weyant, G. Blanford, V. Krey, L. Clarke, J. Edmonds, et al., The
role of technology for achieving climate policy objectives: overview of the
EMF 27 study on global technology and climate policy strategies, Clim. Change
123 (3) (2014) 353e367, http://dx.doi.org/10.1007/s10584-013-0953-7.
 E. Kriegler, K. Riahi, N. Bauer, V.J. Schwanitz, N. Petermann, V. Bosetti, et al.,
Making or breaking climate targets: the AMPERE study on staged accession
scenarios for climate policy, Technol. Forecast. Soc. Change 90 (A) (2015)
 D.J. Immerzeel, P.A. Verweij, F. van der Hilst, A.P.C. Faaij, Biodiversity impacts
of bioenergy crop production: a state-of-the-art review, GCB Bioenergy 6
(2014) 183e209, http://dx.doi.org/10.1111/gcbb.12067.
 M. Labriet, A. Kanudia, R. Loulou, Climate mitigation under an uncertain
technology future: a TIAM-World analysis, Energy Econ. 34 (S3) (2012)
 R. Loulou, A. Lehtilâ‚¬
a, Stochastic Programming and Tradeoff Analysis in TIMES.
Version 3.3 User Note, 2012. Retrieved from the Energy Technology Systems
Analysis Program (ETSAP) website, http://www.iea-etsap.org/web/
 R. Loulou, U. Remme, A. Lehtila, A. Kanudia, G. Goldstein, Documentation for â‚¬
the TIMES Model, 2005. Retrieved from the Energy Technology Systems
Analysis Program (ETSAP) website, http://www.iea-etsap.org/web/
 McKinsey, Reducing U.S. Greenhouse Gas Emissions: How Much at What
Cost?, McKinsey & Company, 2007.
 J. McLaren Loring, Wind energy planning in England, Wales and Denmark:
factors influencing project success, Energy Policy 35 (2007) 2648e2660,
 Mott MacDonald, UK Electricity Generation Costs Update. Brighton, 2010.
 L. Mundaca, L. Neij, E. Worrell, M. McNeil, Evaluating energy efficiency policies with energy-economy models, Annu. Rev. Environ. Resour. 35 (1) (2010)
 National Grid, Solar PV (Electricity Systems) and the National Grid: a Briefing
Note for DECC, 2012. Retrieved from, https://www.gov.uk/government/
 ONS, National Population Projections, 2012-based Projections, Office for National Statistics, Newport, 2013. Retrieved from, http://www.ons.gov.uk/.
 R. Perveen, N. Kishor, S.R. Mohanty, Off-shore wind farm development: present status and challenges, Renew. Sustain. Energy Rev. 29 (2014) 780e792,
 S. Pfenninger, J. Keirstead, Renewables, nuclear, or fossil fuels? Scenarios for
Great Britain’s power system considering costs, emissions and energy security, Appl. Energy 152 (2015) 83e93, http://dx.doi.org/10.1016/
 W. Poortinga, N.F. Pidgeon, S. Capstick, M. Aoyagi, Public Attitudes to Nuclear
Power and Climate Change in Britain Two Years after the Fukushima Accident,
Working Paper UKERC/WP/ES/2013/006, UK Energy Research Centre, 2013.
 S. Pye, N. Sabio, N. Strachan, An integrated systematic analysis of uncertainties
in UK energy transition pathways, Energy Policy 87 (2015) 673e684, http://
 S. Pye, W. Usher, N. Strachan, The uncertain but critical role of demand
reduction in meeting long-term energy decarbonisation targets, Energy Policy
73 (2014) 575e586, http://dx.doi.org/10.1016/j.enpol.2014.05.025.
 Raconteur, UK Wind & Marine Business Barometer 2014, 2014, November 11.
Retrieved from, http://raconteur.net/.
 P. Raskin, F. Monks, T. Ribeiro, D. van Vuuren, M. Zurek, Global scenarios in
historical perspective, in: C. Carpenter, P. Pingali, E. Bennett, M. Zurek (Eds.),
Ecosystems and Human Well-being: Scenarios e Findings of the Scenarios
Working Group Millennium Ecosystem Assessment Series, Island Press,
Washington, DC, 2005, pp. 35e44.
 renewableUK, Wind Energy in the UK – State of the Industry Report 2015,
RenewableUK, London, 2015.
 renewableUK, Offshore Wind Project Timelines, 2015. Retrieved from, http://
 A. Saltelli, P. Annoni, B. D’Hombres, How to avoid a perfunctory sensitivity
analysis, Procedia – Soc. Behav. Sci. 2 (6) (2010) 7592e7594, http://dx.doi.org/
 P. Seljom, A. Tomasgard, Short-term uncertainty in long-term energy system
models – a case study of wind power in Denmark, Energy Econ. 49 (2015)
 N. Strachan, W. Usher, UK MARKAL Modelling e Examining Decarbonisation
Pathways in the 2020s on the Way to Meeting the 2050 Emissions Target
(Final Report for the Committee on Climate Change), 2010. Retrieved from the
UCL Energy Institute website, http://www.ucl.ac.uk/energy-models/models/
 Neil Strachan, Will Usher, Failure to achieve stringent carbon reduction targets in a second-best policy world, Clim. Change 113 (2) (2011) 121e139,
 G. Strbac, M. Aunedi, D. Pudjianto, F. Teng, P. Djapic, R. Druce, A. Carmel,
K. Borkowski, Value of Flexibility in a Decarbonised Grid and System Externalities of Low-Carbon Generation Technologies, Report for the Committee on
Climate Change, Imperial College London & NERA Economic Consulting, 2015.
 S. Syri, A. Lehtila, T. Ekholm, I. Savolainen, H. Holttinen, E. Peltola, Global â‚¬
energy and emissions scenarios for effective climate change mitigationdDeterministic and stochastic scenarios with the TIAM model, Int. J.
Greenh. Gas Control 2 (2) (2008) 274e285, http://dx.doi.org/10.1016/
 M. Tavoni, E. Kriegler, T. Aboumaboub, K.V. Calvin, G. De Maeure, T. Kober, et
al., The distribution of the major economies’ effort in the Durban platform
scenarios, Clim. Change Econ. 4 (4) (2013), http://dx.doi.org/10.1142/
 E. Trutnevyte, EXPANSE methodology for evaluating the economic potential of
renewable energy from an energy mix perspective, Appl. Energy 111 (2013)
 E. Trutnevyte, C. Guivarch, R. Lempert, N. Strachan, Reinvigorating the scenario technique to expand uncertainty consideration, Clim. Change 135 (3)
(2016) 373e379, http://dx.doi.org/10.1007/s10584-015-1585-x.
 Tyndall Centre for Climate Change Research, A Review of Research Relevant to
B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168 167
New Build Nuclear Power Plants in the UK, University of Manchester, 2013.
 W. Usher, The Value of Learning about Critical Energy System Uncertainties,
Doctoral thesis, University College London, 2016. Retrieved from, http://
 W. Usher, N. Strachan, Critical mid-term uncertainties in long-term decarbonisation pathways, Energy Policy 41 (2012) 433e444, http://dx.doi.org/
 W. Usher, N. Strachan, An expert elicitation of climate, energy and economic
uncertainties, Energy Policy 61 (2013) 811e821, http://dx.doi.org/10.1016/
 D. van Vuuren, B. de Vries, A. Beusen, P. Heuberger, Conditional probabilistic
estimates of 21st century greenhouse gas emissions based on the storylines of
the IPCC-SRES scenarios, Glob. Environ. Change 18 (4) (2008) 635e654, http://
 P. Voll, M. Jennings, M. Hennen, N. Shah, A. Bardow, The optimum is not
enough: a near-optimal solution paradigm for energy systems synthesis,
Energy 82 (2015) 446e456, http://dx.doi.org/10.1016/j.energy.2015.01.055.
 Ch Yang, S. Yeh, S. Zakerinia, K. Ramea, D. McCollum, Achieving California’s
80% greenhouse gas reduction target in 2050: technology, policy and scenario
analysis using CA-TIMES energy economic systems model, Energy Policy 77
(2015) 118e130, http://dx.doi.org/10.1016/j.enpol.2014.12.006.
 H. Wainwright, S. Finsterle, Y. Jung, Q. Zhou, J. Birkholzer, Making sense of
global sensitivity analyses, Comput. Geosciences 65 (2014) 84e94, http://
 J. Watson, F. Kern, M. Gross, R. Gross, P. Heptonstall, F. Jones, et al., Carbon
Capture and Storage: Realising the Potential?, Project Report, UK Energy
Research Centre, London, 2012.
 J.P. Weyant, Costs of reducing global carbon emissions, J. Econ. Perspect. 7 (4)
 J.P. Weyant, E. Kriegler, Preface and introduction to EMF 27, Clim. Change 123
(3) (2014) 345e352, http://dx.doi.org/10.1007/s10584-014-1102-7.
 C. Wilson, A. GrÃ¼bler, N. Bauer, V. Krey, K. Riahi, Future capacity growth of
energy technologies: are scenarios consistent with historical evidence? Clim.
Change 118 (2) (2013) 381e395, http://dx.doi.org/10.1007/s10584-012-0618-
168 B. Fais et al. / Energy Strategy Reviews 13-14 (2016) 154e168
Are you busy and do not have time to handle your assignment? Are you scared that your paper will not make the grade? Do you have responsibilities that may hinder you from turning in your assignment on time? Are you tired and can barely handle your assignment? Are your grades inconsistent?
Whichever your reason is, it is valid! You can get professional academic help from our service at affordable rates. We have a team of professional academic writers who can handle all your assignments.
Students barely have time to read. We got you! Have your literature essay or book review written without having the hassle of reading the book. You can get your literature paper custom-written for you by our literature specialists.
Do you struggle with finance? No need to torture yourself if finance is not your cup of tea. You can order your finance paper from our academic writing service and get 100% original work from competent finance experts.
While psychology may be an interesting subject, you may lack sufficient time to handle your assignments. Don’t despair; by using our academic writing service, you can be assured of perfect grades. Moreover, your grades will be consistent.
Engineering is quite a demanding subject. Students face a lot of pressure and barely have enough time to do what they love to do. Our academic writing service got you covered! Our engineering specialists follow the paper instructions and ensure timely delivery of the paper.
In the nursing course, you may have difficulties with literature reviews, annotated bibliographies, critical essays, and other assignments. Our nursing assignment writers will offer you professional nursing paper help at low prices.
Truth be told, sociology papers can be quite exhausting. Our academic writing service relieves you of fatigue, pressure, and stress. You can relax and have peace of mind as our academic writers handle your sociology assignment.
We take pride in having some of the best business writers in the industry. Our business writers have a lot of experience in the field. They are reliable, and you can be assured of a high-grade paper. They are able to handle business papers of any subject, length, deadline, and difficulty!
We boast of having some of the most experienced statistics experts in the industry. Our statistics experts have diverse skills, expertise, and knowledge to handle any kind of assignment. They have access to all kinds of software to get your assignment done.
Writing a law essay may prove to be an insurmountable obstacle, especially when you need to know the peculiarities of the legislative framework. Take advantage of our top-notch law specialists and get superb grades and 100% satisfaction.
We have highlighted some of the most popular subjects we handle above. Those are just a tip of the iceberg. We deal in all academic disciplines since our writers are as diverse. They have been drawn from across all disciplines, and orders are assigned to those writers believed to be the best in the field. In a nutshell, there is no task we cannot handle; all you need to do is place your order with us. As long as your instructions are clear, just trust we shall deliver irrespective of the discipline.
Our essay writers are graduates with bachelor's, masters, Ph.D., and doctorate degrees in various subjects. The minimum requirement to be an essay writer with our essay writing service is to have a college degree. All our academic writers have a minimum of two years of academic writing. We have a stringent recruitment process to ensure that we get only the most competent essay writers in the industry. We also ensure that the writers are handsomely compensated for their value. The majority of our writers are native English speakers. As such, the fluency of language and grammar is impeccable.
There is a very low likelihood that you won’t like the paper.
Not at all. All papers are written from scratch. There is no way your tutor or instructor will realize that you did not write the paper yourself. In fact, we recommend using our assignment help services for consistent results.
We check all papers for plagiarism before we submit them. We use powerful plagiarism checking software such as SafeAssign, LopesWrite, and Turnitin. We also upload the plagiarism report so that you can review it. We understand that plagiarism is academic suicide. We would not take the risk of submitting plagiarized work and jeopardize your academic journey. Furthermore, we do not sell or use prewritten papers, and each paper is written from scratch.
You determine when you get the paper by setting the deadline when placing the order. All papers are delivered within the deadline. We are well aware that we operate in a time-sensitive industry. As such, we have laid out strategies to ensure that the client receives the paper on time and they never miss the deadline. We understand that papers that are submitted late have some points deducted. We do not want you to miss any points due to late submission. We work on beating deadlines by huge margins in order to ensure that you have ample time to review the paper before you submit it.
We have a privacy and confidentiality policy that guides our work. We NEVER share any customer information with third parties. Noone will ever know that you used our assignment help services. It’s only between you and us. We are bound by our policies to protect the customer’s identity and information. All your information, such as your names, phone number, email, order information, and so on, are protected. We have robust security systems that ensure that your data is protected. Hacking our systems is close to impossible, and it has never happened.
You fill all the paper instructions in the order form. Make sure you include all the helpful materials so that our academic writers can deliver the perfect paper. It will also help to eliminate unnecessary revisions.
Proceed to pay for the paper so that it can be assigned to one of our expert academic writers. The paper subject is matched with the writer’s area of specialization.
You communicate with the writer and know about the progress of the paper. The client can ask the writer for drafts of the paper. The client can upload extra material and include additional instructions from the lecturer. Receive a paper.
The paper is sent to your email and uploaded to your personal account. You also get a plagiarism report attached to your paper.
Delivering a high-quality product at a reasonable price is not enough anymore.
That’s why we have developed 5 beneficial guarantees that will make your experience with our service enjoyable, easy, and safe.
You have to be 100% sure of the quality of your product to give a money-back guarantee. This describes us perfectly. Make sure that this guarantee is totally transparent.Read more
Each paper is composed from scratch, according to your instructions. It is then checked by our plagiarism-detection software. There is no gap where plagiarism could squeeze in.Read more
Thanks to our free revisions, there is no way for you to be unsatisfied. We will work on your paper until you are completely happy with the result.Read more
Your email is safe, as we store it according to international data protection rules. Your bank details are secure, as we use only reliable payment systems.Read more
By sending us your money, you buy the service we provide. Check out our terms and conditions if you prefer business talks to be laid out in official language.Read more